Beta-Thalassemia is one of the commonest inherited diseases in humans, characterized by a severe hemolytic anemia and ineffective erythropoiesis. While transfusion and chelation therapy do represent a radical treatment, the use of bone marrow replacement is limited is limited by complications of allogeneic transplantation and the need for aggressive conditioning regimens. Thus, the goal of this proposal is to develop a treatment that integrates a genetic correction in autologous hematopoietic stem cells (HSC) with a reasonable transplantation strategy. The approach we propose is based on efficient lentiviral- mediated transfer of the wild-type beta-globin gene in cord blood or peripheral blood stem cells, together with a strategy for selection of genetically modified cells that are applied in vivo after transplantation. Our recent results establish that efficient gene transfer of a modified beta- globin gene and large elements of the beta-globin locus control region (LCR) can be achieved using recombinant lentiviruses. We have demonstrated that successful incorporation of large LCR elements increases mean beta-globin expression to therapeutically relevant levels. The major goals of this project are: (a) to improve erythroid-specific gene expression from a virally encoded beta-globin transcription unit; (b) to investigate whether thalassemia intermedia and thalassemia major can be cured in animal models of disease, expanding on our recent demonstration that we can achieve therapeutically relevant levels of human beta-globin in long-term murine bone marrow chimeras; (c) to confer a competitive advantage to the transduced HSC for repopulation of the host marrow using resistance to methotrexate as a model; and (d) to test our vectors in human hematopoietic cells. we propose a detailed analysis of the function of the LCR and of the chicken globin insulator (which we recently showed increases the probability of retroviral expression at random integration sites and decreases vector silencing) in stringent in vitro and in vivo assays that are relevant to the critical evaluation of their therapeutic potential. These studies include investigations in murine models of beta-thalassemia and in primary human CD34+ cells of normal subjects and patients. To analyze globin gene expression and the gene, we will capitalize on our ability to efficiently derive erythroid progeny from long-term cultured CD34+ cells and our experience with mouse/human xenochimeras using NOD-scid/scid mice. We ultimately aim to establish by direct experimental evidence that expression of the lentivirus-encoded human globin gene is sustained over time in murine and human cells in vivo and that adequate expression of the mutant dihydrofolate reductase permits efficient in vivo selection with methotrexate-trimetrexate.